The use of dried blood spots (DBS) which has successfully been introduced in neonatal metabolic screening is an appropriate method to reduce virus infection risk to a minimum, facilitating regular mailing and handling of samples in the laboratory. Injection diacetylmorphine use is notably associated with a prevalence of infection and a risk of transmission of blood-borne viruses. The aim of the present study was to establish a method to determine morphine and 6-acetylmorphine (6-AM) as accurately and sensitively from DBS as from whole blood. Analysis by liquid chromatography/tandem mass spectrometry was checked for carryover, ion suppression/enhancement, linearity of response, lower limits of detection and quantification, and the within-run and between-run assay imprecision for both whole blood and DBS after liquid/liquid extraction. DBS drying time and elution were optimized, and extraction efficiency from DBS was compared with that of blood and of a solution of the pure compounds. Short-term stability of morphine and 6-AM was determined at −20°C, 4°C, and 40°C up to 7 days from both whole blood and DBS. Furthermore, it was tested whether analysis of DBS may be as reliable as that of whole blood investigating 50 authentic samples. The lower limit of detection was 0.4 ng of morphine per spot and 0.8 ng of 6-AM per spot using a DBS blood volume of 100 μL and was 0.3 and 0.7 ng/100 μL whole blood for morphine and 6-AM, respectively. Recovery rates of the analytes from DBS did not differ from those from whole blood and were ≥55% for 6-AM and ≥25% for morphine, and the within- and between-run coefficients of variation were always ≤9%. 6-AM degraded rapidly at both 4°C and 40°C in whole blood; however, it seemed to be much more stable in DBS. Significant correlations between real whole-blood samples and DBS were obtained. The DBS assay has potential as a precise and inexpensive option for the determination of morphine and 6-AM in small blood samples. Further, the DBS matrix proved to excellently stabilize 6-AM.